transhumanist-already-exists/karpotron-tokenizer

Karpotron Tokenizer

Overview

Karpotron is a specialized tokenizer based on NVIDIA's Nemotron-3-Nano-30B tokenizer, optimized for Ukrainian language processing. It replaces 28,065 vocabulary slots with Ukrainian/Cyrillic BPE entries while maintaining the original 131,072 vocabulary size by pruning tokens from writing systems geographically and culturally distant from Ukraine.

Key Features

1. +28,065 Cyrillic BPE entries:

   • 152 base Cyrillic letters
   • 27,425 Ukrainian surface tokens from lapa-llm/tokenizer
   • 488 reachable Ukrainian replacement/intermediate nodes used to repair strict BPE reachability

2. No removal of English or EU language tokens - only non-essential tokens from distant writing systems were replaced

3. Latin-safe removal - preserved Latin tokens with diacritics used in multiple European languages (ü, ö, ç, ã, á, é, etc.)

4. Identical specifications - vocab size (131,072), byte-level BPE encoding match original Nemotron

Reachability Repair

The tokenizer was repaired with a strict byte-level BPE reachability audit over the replaced ID range (id > 103006) and then checked against the actual tokenizers ByteLevel pre-tokenizer + BPE pipeline. The current artifact passes:

Checked inserted entries	Atomic entries	Broken entries
28,065	28,065	0

During repair, 488 previously unreachable or pipeline-unreachable surface slots were replaced with reachable Ukrainian replacement/intermediate nodes while keeping the vocabulary size and ID range unchanged. See reachability_repair_report.md and reachability_actual_tokenizers_report.md for examples.

Replaced Tokens by Writing System

The tokenizer replaced 24,360 tokens from 17 writing systems while adding 28,065 Ukrainian tokens:

Writing System	Original	Removed	Retained	% Removed
Arabic	9,400	8,458	942	90.0%
Hangul (Korean)	4,492	3,330	1,162	74.1%
Han (Chinese)	3,767	2,364	1,403	62.8%
Devanagari (Hindi)	1,554	1,155	399	74.3%
Hebrew	1,002	722	280	72.1%
Telugu	920	675	245	73.4%
Bengali	839	615	224	73.3%
Armenian	1,121	549	572	49.0%
Thai	567	423	144	74.6%
Kannada	570	419	151	73.5%
Tamil	539	394	145	73.1%
Malayalam	406	283	123	69.7%
Georgian	513	240	273	46.8%
Hiragana/Katakana (Japanese)	1,623	208	1,415	12.8%
Gujarati	204	136	68	66.7%
Gurmukhi	155	111	44	71.6%
Myanmar	234	96	138	41.0%

Fully preserved:

• Latin scripts (English, Spanish, French, German, Italian, Portuguese, Dutch, Danish, Swedish, Polish)
• Greek (1,507 tokens, 100% retained)

Metrics

Acknowledgement: evaluation results provided by @Sofetory

| |lang-uk/malyuk [100k] |allenai/c4(en) [100k]|allenai/c4 (es,fr,it,de) [100k]|QIRIM/crh (Cyrillic) [94]|allenai/c4(ru) [100k]|allenai/c4(bg) [100k] |allenai/c4(be) [100k]| |words count |22,898,164 |36,170,971 |198,173,216 |1,868,259 |42,557,519 |44,627,199 |43,153,645 | | | | | | | | | | |tokenizers |tokens |toks/word |tokens |toks/word |tokens |toks/word |tokens |toks/word |tokens |toks/word |tokens |toks/word |tokens |toks/word | |Qwen/Qwen3-8B |84,408,084 |3.686 |46,884,593 |1.296 |395,581,536 |1.996 |7,956,741 |4.259 |116,115,062 |2.728 |132,597,427 |2.971 |173,571,099 |4.022 | |meta-llama/Llama-3.1-8B-Instruct |57,226,997 |2.499 |46,085,724 |1.274 |382,143,751 |1.928 |7,386,873 |3.954 |104,974,733 |2.467 |119,123,733 |2.669 |150,189,294 |3.48 | |microsoft/Phi-4-mini-instruct |59,447,036 |2.596 |45,423,925 |1.256 |335,188,687 |1.691 |5,995,822 |3.209 |91,824,464 |2.158 |102,472,523 |2.296 |119,587,038 |2.771 | |CohereLabs/aya-expanse-8b |50,973,632 |2.226 |47,364,187 |1.309 |353,221,932 |1.782 |6,614,719 |3.541 |93,089,697 |2.187 |112,612,668 |2.523 |141,262,943 |3.273 | |google/gemma-3-12b-it |57,388,402 |2.506 |47,285,432 |1.307 |354,241,840 |1.788 |6,240,944 |3.341 |95,520,817 |2.245 |103,950,626 |2.329 |131,398,147 |3.045 | |nvidia/NVIDIA-Nemotron-3-Nano-30B|62,087,149 |2.711 |47,630,139 |1.317 |365,218,644 |1.843 |6,623,516 |3.545 |107,233,038 |2.520 |108,691,963 |2.436 |135,489,439 |3.140 | |karpotron-tokenizer (Ours) |46,456,626 |2.029 |47,650,584 |1.317 |365,285,307 |1.843 |7,519,362 |4.025 |132,519,787 |3.114 |131,626,936 |2.949 |158,657,784 |3.677 | |Comments |25% fewer tokens than Nemotron on Ukrainian|English unchanged|EU languages unchanged|QIRIM slightly worse|Russian drops (UA-centric)|Bulgarian drops slightly|Belarusian drops|

Usage Example

from transformers import AutoTokenizer

tokenizer = AutoTokenizer.from_pretrained(
    "transhumanist-already-exists/karpotron-tokenizer"
)

toks = tokenizer("Всі красиві зберігають оптимізм", add_special_tokens=False)
print(toks.input_ids)  # 6 tokens with the current reachability-repaired artifact

Model Contents

• tokenizer.json - Byte-level tokenizer spec (131,072 tokens, 258,350 merges)
• tokenizer_config.json - Configuration metadata
• special_tokens_map.json - Special token mappings (identical to Nemotron)
• reachability_repair_report.md/json - Strict BPE graph reachability repair summary
• reachability_actual_tokenizers_report.md/json - Actual tokenizers pipeline reachability repair summary

Embedding Initialization

For newly added tokens in Nemotron models, you can:

• Use tools like Focus or Zett
• Initialize embeddings randomly with warm-up schedule training
• Unchanged tokens (103,007) retain original IDs and can reuse existing embeddings

Citation

@misc{karpotron_tokenizer,
  author       = {Bohdan Didenko},
  title        = {Karpotron Tokenizer},
  year         = {2026},
  howpublished = {\url{https://huggingface.co/transhumanist-already-exists/karpotron-tokenizer}},
  note         = {Based on NVIDIA Nemotron-3-Nano-30B and lapa-llm/tokenizer}
}

Base Models

• Base: nvidia/NVIDIA-Nemotron-3-Nano-30B-A3B-Base-BF16
• Donor: lapa-llm/tokenizer